Using the SDK approach you take the set of instance ids embedded in the response/result data returned from the DescribeAutoScalingInstances call and pass them to the DescribeInstances call for EC2 (using new-object again to get an EC2 client and request objects). This will net you a collection of Amazon.EC2.Model.Reservation objects (again inside the response/result data) from which the RunningInstance collection inside each reservation will get you the ip address(es) for the EC2 instance.
It is however much simpler to use the AWS Tools for Windows PowerShell like this:
Get-ASAutoScalingInstance | select -expandproperty InstanceId | Get-EC2Instance | select -expandproperty RunningInstance | ft InstanceId, IpAddress
Get-ASAutoScalingInstance maps to the request in your question; this yields the set of EC2 instances from which we extract the id of each instance with a select. We then request details for the instance using Get-EC2Instance; as noted above this yields an Amazon.EC2.Model.Reservation object, within with are the details of the instance (in the RunningInstance collection). We flatten this to pull out the instance id and associated ip address for the table.
The sample pipeline above assumes you've set credentials and region to use in the shell using Set-AWSCredentials and Set-DefaultAWSRegion.
The AWS Tools for Windows PowerShell are included with the download msi for the SDK and Visual Studio toolkit available here http://aws.amazon.com/net/.
Hope this helps.